/*
Timothy Ryan	
Proposal title: Are losers gullible? A new test of ideological asymmetry in conspiracy beliefs


HYPOTHESES

Stated-Hyp1: Is credulity assymetric?

	Test-Hyp1: Republicans are more likely than Democrats to believe a statement that they have no priors for.

********************************************************************************
NOTES:
The research question is whether conservatives are more likely to believe in 
conspiracy theories because right-wing theories have been more believable, or 
if conservatives are just more gullible. No hypotheses were stated. 

*/

clear all
use "RyanS14.dta", clear

********************************************************************************

* INDICATORS OF EXPERIMENTAL MANIPULATION

* Treatment variable 
	/* 3 statements given:
	Did the [INSERT: Q2_INSERT] legislators in Ohio accept laundered money from a group of Canadian steel manufacturers, hoping to improve their business dealings in the state?
	Did Lucas Hofmann, a [INSERT: Q3_INSERT] prosecutor in Wisconsin, plot with [INSERT: Q3_INSERT] Party members to suppress evidence that a Gerry Mason, a wealthy donor in the state, engaged in pedophilia?
	Was the price of crude oil higher on March 1, 2016 than on October 1, 2016?
	
	Note: we don't need the treatment var for this analysis
	*/
	

* OUTCOME MEASURES

	forval i=2/4 {
	replace q`i'=. if q`i'>5 // missing values recode
	gen belief`i'=6-q`i' if q`i'!=. // recode so increasing vals indicate greater belief

	tab belief`i', mis
	}
	
	gen belief= belief2+belief3+belief4
	sum belief
 	
* MODERATORS

	tab partyid5
	recode partyid5 (-1=.) (1/2=0) (3=.) (4/5=1), gen(republican)
	tab republican, mis

	
********************************************************************************

* ANALYSIS
	

*Test-Hyp1: Republicans are more likely than Democrats to believe a statement that they have no priors for.

	reg belief i.republican
	tess 1.republican +, init(RyanS14)
		// result: republicans are not different. p-val. 0.242


	
